DESCRIPTION: (Adapted From The Investigator s Abstract): This study focuses on the predictors, precipitants, and sequelae of relationship violence among inner-city late- adolescents. It builds on data from a four-year longitudinal study of the community, family, and peer, risk predictors of antisocial behavior among inner-city adolescent males. By adding two additional interviews with these men, their female partners, and additional females of similar age, ethnicity, and residence location we can trace the antecedents of violence in intimate relationships. Also, an ethnography of situations and conditions that are the catalysts for relationship violence is undertaken to inform the focus of the longitudinal quantitative research. This study focuses on the members of society most at-risk for serious antisocial and violent behavior, premature death due to violence, and with elevated risk for relationship violence. However, this population is vastly understudied and misunderstood. There is a dearth of studies focused on this age period that is critical for developing sound risk and prevention strategies. This study will provide needed multi-level, multi-variate data on the primary determinants of risk, the consistency of risk factors with theoretical contentions, and the consistency across communities, and ethnic and gender groups. Applying an ecological perspective on risk factors and a transactional perspective on risk development, this study can provide needed clarification of outcomes that are the focus of many preventive efforts, specify the salient predisposition and precipitating risk factors and behavioral sequences, and clarify how relationship violence relates to other violence and general antisocial behavior. In addition, it can provide some of the first detailed information on how partners contribute to the formation of violent relationships, how such attitudes and practices may be formed during late adolescence, and the variations of such contributions by gender and across ethnic groups. A sample of 366 families, with 92% average retention from wave to wave, have been followed for four annual interviews with prior funding. At the last wave, in a pilot for this study, 106 boys reported a romantic relation that lasted at least three months and 1/3 of those reported using violence in that relationship. Using snowball sampling methods we will solicit participation by the young men's romantic partners and additional female friends of the partners to obtain parallel data over two years. 40 youth will be followed in an ethnographic study.